Machine Learning

Ready

Regression first

What it means for a model to learn, then the first fit: a line. Scaling and shrinkage keep that line honest.

1. 01Introduction to Machine Learning
2. 02Linear Regression
3. 03Normalization and Regularization

Ready

Data and validation

Missing values, encodings, and correlations before any fit. Then the split that tells you whether the fit is real.

1. 04Data Preprocessing and EDA
2. 05Feature Encoding & Correlation Analysis
3. 06Training, Testing Sets & Cross Validation

Ready

Error and least squares

Where error comes from, how to measure it, and the line that makes the squared version smallest.

1. 07Bias-Variance Tradeoff
2. 08MSE, RMSE, and MAE
3. 09Least Squares Optimization

Ready

Classification and training

A probability instead of a number, the step that trains it, then two more ways to draw a boundary: Bayes and a tree.

1. 10Logistic Regression
2. 11Gradient Descent
3. 12Bayesian Classification
4. 13Decision Trees

Ready

Structure without labels

The piles the data falls into, and the direction it spreads along the most.

1. 14K-Means Clustering
2. 15Principal Component Analysis

Ready

Ensembles, softmax, and losses

Many trees voting, many classes competing, a fit that bends locally, and the losses that drive training.

1. 16Ensemble Methods
2. 17Softmax Regression
3. 18Locally Weighted Regression
4. 19Cost Functions & Cross Entropy

Ready

Margins, neighbors, and metrics

The widest street between two classes, the vote of the nearest points, and the metrics that judge any classifier.

1. 20Support Vector Machines
2. 21K-Nearest Neighbors
3. 22Model Evaluation Metrics

Ready

Neural networks

Stacked layers of the fits you already know: a network, an autoencoder that compresses, the fundamentals, then images and sequences.

1. 23Artificial Neural Networks
2. 24Autoencoders: Unsupervised Representation Learning
3. 25Neural Networks Fundamentals
4. 26Convolutional Neural Networks (CNNs)
5. 27Recurrent Neural Networks (RNNs)